Re: Scalability and oversight (Was: Just in case you're curious)

2003-12-28 Thread Steven Noels
On Dec 27, 2003, at 7:39 PM, Santiago Gala wrote:

Scalable because big groups of people can coordinate, even if they 
don't give specific input or they were not there while the decision 
was taken.
OT: after some light holiday-time reading (Prey from Michael Crichton 
- http://www.amazon.com/exec/obidos/ASIN/0061015725/), it's funny to 
try and invent some parallels between open source software communities 
and the agent swarms outlined in his novel. Freaky.

/Steven
--
Steven Noelshttp://outerthought.org/
Outerthought - Open Source Java  XMLAn Orixo Member
Read my weblog athttp://blogs.cocoondev.org/stevenn/
stevenn at outerthought.orgstevenn at apache.org
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


Re: Scalability and oversight (Was: Just in case you're curious)

2003-12-27 Thread Geir Magnusson Jr.
On Dec 27, 2003, at 1:39 PM, Santiago Gala wrote:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
El lunes, 22 dici, 2003, a las 16:32 Europe/Madrid, Geir Magnusson Jr. 
escribió:

You are free to do what you want.  Is this then about personal google 
hitcount?

To the risk of re-starting a extinguishing discussion, I think google 
(or any outsider looking) plays an important role here, but not in the 
personal hitcount sense.
I'll simply note that as you didn't quote what I was responding to, 
some readers unfamiliar with the thread might incorrectly assume that 
this was about an effort to keep this from being an open discussion.

No one wanted to keep this from being an open discussion.  It was first 
suggested by Peter a while ago, and I think everyone was in agreement.  
The issue was trying to get some organization and planning around a 
complicated subject before bringing it public.

I think openness of product *and* process is the only thing that makes 
us scalable and fault-tolerant, when comparing Apache with more 
traditional organizations.
I fully support openness, but I'll also note that a bit of organization 
and planning go a long way.  And there are plenty of traditional closed 
organizations that do just fine due to planning and organization, such 
as IBM and Microsoft.

Scalable because big groups of people can coordinate, even if they 
don't give specific input or they were not there while the decision 
was taken.
Yep, all helped by a bit of planning and organization.

Fault tolerant because the public audit trail left in CVS and mailing 
lists makes it easy for third party observers (or interested parties) 
to spot any error in oversight.
Yep, all helped by a bit of planning and organization.  Note that 'CVS' 
and 'mailing lists' are two examples of planning and organization.

If we go to the cathedral versus bazaar metaphor, nothing beyond a 
small group conversation remains private in the bazaar. So, if some 
merchant down there is selling cheaper, notice propagates fast. Same 
if some merchandise is faulted.
Maybe.  I'll note that the most successful OSS projects I've seen also 
had a strong individual or group of individuals that helped via (you 
can guess what's coming...), ...a bit of planning and organization.  
Apache httpd, linux, emacs, hibernate, mysql, the list goes on...

Same w/ Jakarta.  There have always been a strong group of people 
guiding the sub-projects and the project overall.  What we are trying 
to do now is increase that group, or better, recognize those that are 
doing it already, and conforming to legal structure needed by the ASF.

geir

--
Geir Magnusson Jr   203-247-1713(m)
[EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]